977-fix-fractals-documentation

topic	governance
type	audit
status	research-complete
last-validated	2026-07-06
superseded-by
related-docs	975, 718, 696, 705, 114, 102
original-query	Fix fractals documentation - the fractal docs are stale/wrong in places; catalogue what is wrong vs the verified on-chain facts and produce a fix-list
tier	STANDARD

977 - Fix the fractals documentation (correction catalogue)

Goal: The ZAO fractal/governance docs carry numbers that the 2026-07-05 on-chain pull proved wrong. This doc is the exact fix-list: every stale claim, its file and line, and the correction - so the whitepaper foundations and lineage docs can be brought to ground truth in one pass. Doc 975 is the citable source for every correction here.

Why now

Doc 975 pinned the live on-chain numbers for ZAO Respect. Several older governance docs were written from modeled estimates (doc 718b) and now contradict the measured reality. Contradictory docs are worse than no docs - the whitepaper is Zaal's "magnum opus" and its foundations must not cite figures that a reader can disprove on Blockscout in two minutes.

The corrections (grounded, file:line)

Every "should be" below traces to [[reference_zao_respect_onchain_facts]] (on-chain pull 2026-07-05) via Doc 975.

1. OREC vote/veto windows: 48h -> 72h

File	Line(s)	Says	Should say
718c-ordao-onchain-architecture.md	22, 43, 51, 236, 409	"48h voting + 48h veto", "48-hour veto window = 86,400s"	72h each (voteLen=vetoLen=259,200s on-chain). The 86,400s math is also wrong - 72h = 259,200s
705-fractal-governance-external-deep-research/README.md	245-246	"Voting 48 hours / Veto 48 hours"	72 hours each

2. Proposal threshold: "10% of Respect" -> ~2.6% (1,000 Respect)

File	Line	Says	Should say
718c-ordao-onchain-architecture.md	64, 409	"e.g. 10% of total Respect"	minWeight = 1,000 Respect = ~2.6% of OG supply (38,484)
705-...deep-research/README.md	247	"5-10% of total Respect... 10% as typical framing"	~2.6% measured; drop the "10% typical" framing

3. Distribution Gini: 0.23 -> 0.73 (and the nuance)

File	Line	Says	Should say
696-respect-fractal-lineage-summary/README.md	129	"a single Fibonacci round produces a Gini around 0.23"	The cumulative OG Gini is ~0.73 (top 10 = 53%). A single 6-person Fibonacci round computes to 0.41, not 0.23. The "dramatically more equal" claim is overstated - keep the comparison honest

This is the most important correction: the docs currently claim the distribution is near-egalitarian (0.23). On-chain it is meaningfully concentrated (0.73). The whitepaper's credibility rests on stating this honestly (per the whitepaper memory, Chapter 9 Limitations must stay honest).

4. Streak: "90+ weeks" -> "100+ weeks" (~101)

File	Line	Says	Should say
696-...lineage-summary/README.md	33, 169	"90+ weeks" (x2)	~101 weeks ("100+ weeks")
114-zao-fractal-live-infrastructure/README.md	350	"90+ weeks"	~101 weeks
102-fractals-frapps-ordao-page/README.md	94	"90+ weeks running"	~101 weeks

Understated, not wrong - but "100+ weeks" is a stronger, true claim. Update it everywhere.

5. Member count: ~200 -> 156

Wherever any fractal doc says "~200 members / ~200 holders", correct to 156 unique Respect holders (122 OG + 55 ZOR, 21 both). The ~200 figure conflated Discord/community size with on-chain holders. (Not all instances grepped here - the site copy fix in doc 975's action table covers the user-facing surfaces; this doc covers the research docs.)

What is NOT a documented error (leave alone)

• The mechanism descriptions (three-phase optimistic OREC, Fibonacci consensus, weekly cadence, ZOR/OG two-token split) are correct - only the numeric parameters drifted. Do not rewrite the prose, just fix the numbers.
• The combined OG+ZOR Gini is deliberately not published (ZOR ERC-1155 encoding is ambiguous - see doc 975). Any doc that wants a combined figure must wait; publishing OG-only 0.73 labeled as such is the correct interim.

Also See

• Doc 975 - the citable source for every correction above; link it from each doc as the numbers get fixed.
• [[reference_zao_respect_onchain_facts]] - the raw 2026-07-05 on-chain pull.
• Doc 718 - the whitepaper foundations hub; 718b (modeled) is the origin of most stale figures and should be marked superseded-by: 975 for its numbers.

Next Actions

Action	Owner	Type	By When
Edit 718c lines 22/43/51/64/236/409: 48h->72h, fix the 86,400s math to 259,200s, 10%->~2.6% (1,000 Respect)	@Zaal	PR	2026-07-13
Edit 705 lines 245-247: 48h->72h (both), threshold ->~2.6%	@Zaal	PR	2026-07-13
Edit 696 line 129: Gini 0.23 -> 0.73 cumulative (single round = 0.41), soften "dramatically more equal"	@Zaal	PR	2026-07-13
Edit 696/114/102: "90+ weeks" -> "100+ weeks (~101)" everywhere	@Zaal	PR	2026-07-13
Add superseded-by: 975 to doc 718b frontmatter (its numbers, not its structure)	@Zaal	Edit	2026-07-13

Sources

• [FULL] 718c-ordao-onchain-architecture.md, 705-.../README.md, 696-.../README.md, 114-.../README.md, 102-.../README.md (ZAOOS working tree, grepped 2026-07-06) - the exact stale lines are cited above.
• [FULL] Doc 975 + reference_zao_respect_onchain_facts - the measured corrections.
• [FULL] project_zao_fractal_whitepaper memory - confirms the whitepaper is the flagship doc and Chapter 9 Limitations must stay honest, which is why the Gini correction matters most.